SYSTEM DESIGN — NEWFOUNDLAND-NOVA SCOTIA LINK 241 



and on ship, portable injection-moulding machines and X-ray equip- 

 ment being specially designed for handling over the bog. A single pair 

 cable was also laid in the main cable trench to provide speaker facilities 

 between Clarenville and Terrence\'ille (which has no public telephone), 

 with intermediate positions for use of the lineman. As a measure of pro- 

 tection against lightning strikes, two bare copper wires were buried 

 about 12 inches apart and 6 inches above the cable. Both the construc- 

 tional work in Newfoundland"* and the laying operation at sea^ have been 

 described elsewhere. 



TEST RESULTS 



In the interval between the completion of the link in May, 1956, and 

 its incorporation in the transatlantic system, tests were carried out to 

 establish its performance and day-to-day variations; an assessment of 

 the annual variations has, of course, been impossible at this date. 



Variation of Transmission Loss 



Close observation of the transmission loss of the 92 kc pilots on 

 Groups 1 and 5 leads to the following tentative conclusions: 



(a) Over periods of 1 hour the variations are not measurable, i.e., 

 less than =b0.05 db. 



(6) Over periods of 24 hours there are no systematic changes; ap- 

 parently random changes of about 0.1 db are probably attributable to 

 the measuring equipment. 



(c) Over a period of eight weeks (July and August, 1956) there was 

 a systematic increase in loss of about 0.3 db. By means of the loop-gain 

 equipment it has been possible to deduce that most of this change has 

 occurred in the land section. 



The results indicate that the submarine cable link has better day-to- 

 day stability than the best testing equipment which it has been possible 

 to provide. Many more data will clearly be necessary befoi'e the annual 

 variations can be definitely established, but the present indications 

 are that these will be less than those assumed in the design of the link. 



A ttenuation/ Frequency Characteristics 



The frequency characteristics of the supergroup in the two directions 

 of transmission are shown in Fig. 14. It will be seen that in no transat- 

 lantic groups does the deviation from mean exceed ±0.35 db. 



Circuit Noise 



Table II shows the noise level on Channels 1 and 12 of each of the 

 five groups measured without traffic on the system. 



